126 BELL SYSTEM TECHNICAL JOURNAL 



If we may assume that the low melting glasses (1-4) worked with 

 in this investigation received approximately the same heat treatment 

 during their manufacture, then the data presented in Table II would 

 indicate that there is a definite relation between gas evolved and the 

 chemical composition of the glass. The amount of gas given up per 

 square centimeter of glass surface is found to be closely parallel to 

 the alkali content except in the case of glass number 6. This par- 

 allelism holds more closely for the amount of water vapor than it 

 does for the other gases, although the relation does hold to a less 

 striking extent for carbon dioxide. In a paper by Niggli (24), that 

 treats of the phenomena of equilibrium between R 2 0, Si 2 and C0 2 

 (R 2 = Na 2 0, K 2 0, Li 2 0), in melts at temperatures of 900 to 1000°C. 

 under a pressure of one atmosphere C0 2 , it is interesting to note, that 

 the results obtained show the amount of C0 2 in the melt, when equi- 

 librium is reached, to decrease as the composition of the melt becomes 

 less alkaline. The amount of permanent gases evolved is roughly 

 the same for the various types of glass. The reason for glass number 

 6 falling out of line with the other glasses will be discussed later. The 

 authors, after making inquiries at several glass manufacturing con- 

 cerns, feel that the statement can be made with a fair degree of justice 

 that ordinary commercial glasses of like composition usually receive 

 approximately the same heat treatment in their manufacture. Pres- 

 ervation of melting pots, saving in fuel, etc., make it very essential 

 for the glass manufacturer to know the lowest temperature that can 

 be used with the assurance of producing good glass. In all the glass 

 factories where we made inquiries we found that this minimum tem- 

 perature was about constant in cases where glasses having about the 

 same melting points were being made. In these factories we 

 were also told that the furnaces were usually held as near this 

 minimum temperature as possible throughout the entire melting 

 process. 



Glass number 5 undoubtedly received a higher heat treatment in 

 the melting process, because of its higher melting point and greater 

 viscosity, than did the other glasses that were tested. The data 

 given in Table II shows that this glass gave off very little gas when 

 subjected to heat treatment in vacuo. The high heat treatment 

 that this glass received in its manufacture is probably responsible in 

 part for this small evolution of gas, but its low alkali content is prob- 

 ably equally responsible. It will be noted when reference is made to 

 Table II that this glass was heated to 500°C. whereas the other glasses 

 were only subjected to a heat treatment of 400°C. 



